Does copper beam-hardening filtration reduce imaging dose without compromising targeting accuracy in robotic radiosurgery?

OBJECTIVE: Accurate target localization in frameless stereotactic radiosurgery (SRS) is commonly achieved using intrafraction kV X-ray imaging, which contributes to patient imaging dose. In this study, the feasibility of reducing the imaging dose in CyberKnifeTM SRS by implementing copper (Cu) beam-hardening filtration while preserving image-guided targeting accuracy is investigated. APPROACH: Entrance Skin Dose (ESD) at isocenter was measured for standard aluminum (Al) filtration and additional Cu filtration thicknesses of 0.12 mm and 0.24 mm. Monte Carlo simulations were used to characterize Cu-induced X-ray spectral changes and to estimate imaging doses using patient-specific voxelized anatomical models over a range of tube potentials and filtration conditions. The effect of Cu filtration on targeting accuracy was assessed using End-to-End (E2E) tests, supplemented by patient image data. MAIN RESULTS: Copper filtration progressively removed low-energy photons, increasing the mean photon energy from 56 keV with standard Al filtration to 60 keV and 64 keV for 0.1 mm and 0.2 mm Cu filtration at 120 kVp. At 10 mAs, ESD increased from 0.248 mGy at 80 kVp to 0.822 mGy at 150 kVp with Al filtration, while addition of 0.12 mm and 0.24 mm Cu reduced ESD by 46% and 64% at 80 kVp and by 25% and 36% at 150 kVp, respectively. Patient model calculations confirmed meaningful organ dose sparing for the eye lenses, thyroid gland, and skin. For 6Dskull tracking, implementation of Cu filters induced modest image brightness/gradient error changes which were readily compensated by small kVp increase without mAs adjustment. For Xsight Spine and fiducial tracking algorithms registration performance was maintained with the addition of Cu filtration. E2E testing confirmed that the total system targeting error was maintained for all three tracking algorithms studied. SIGNIFICANCE: Copper filtered imaging constitutes an effective strategy for optimizing imaging dose in the evaluated CyberKnife tracking workflows without compromising tracking accuracy.